Development Of A Toolset For Modeling The Urban Traffic Flow With Influence Of Pluvial Flooding

Metropolis are complex socio-ecological systems which experience the intense interplay between physical conditions and human activities.However,in the face of climate change and rapid urbanization,most urban areas have become very vulnerable to pluvial flooding hazard due to an increase of extreme heavy rains and the inefficiency of drainage systems.Particularly,pluvial flooding may pose great threat to the urban traffic network which supports a large amount of cargo movement and travel activities,and thus it can cause considerable inconvenience to residents' livelihoods if there are traffic lines disrupted,or even crippled,as a result of poor visibility,widespread puddles,road closures,etc.Therefore,it is important to study and reveal how pluvial flooding affects the dynamic traffic flow,and how travel behaviors in the traffic network response to the effects of varying rainfall and flooding events.A precise estimation of traffic status in the face of rainfalls can be useful for traffic authorities to manage transportation systems using control,advisory,and road treatment strategies,and to broaden the ability of individuals to make flexible decisions regarding travel mode,route and time.This thesis sets out to propose a novel method for simulating the effects of pluvial flooding on urban traffic,and develop an easy tool to implement dynamic simulation.Then,a case study was conducted for the north Huangpu District of Shanghai.Major contributions and findings of this study are listed below.(1)Generate compound scenarios by incorporating pluvial flooding and various traffic demand.Specifically,in order to generate pluvial flooding of different hazard levels,varying precipitation of three return periods(10-,50-and 100-year)were produced on basis of the Shanghai rainstorm intensity formula and Keifer and Chu method.Then,FloodMap-HydroInundation 2D,a hydrodynamic modeling package,was adopted to simulate the flooding driven by each precipitation scenario.Meanwhile,in the light of the Annual Traffic Report of Shanghai in 2015,and the feedback from consultants in Shanghai Municipal Transportation Commission,traffic demand was set as 6000,8000 and 10000 vehicles per hour,and randomly distributed in the study area.Finally,a total of 12 compound scenarios were created through incorporating 2 weather conditions(sun and rain),3 precipitation magnitudes and 3 traffic demand levels.(2)Propose a novel method for simulating the effects of pluvial flooding on urban traffic.Based on a careful literature review and systematical understanding of how rainfall events affect traffic flow and travel behaviors on different spatial and time frames,this thesis gave its focus on the effects on vehicle speed and the road condition such as blocked or passable.Specifically,as learned from the historical experience,a road segment need to be marked as blocked if the flooding depth exceeded 30 cm.In this manner,outputs of FloodMap-HydroInundation 2D were initially integrated into the road network for subsequent traffic simulation.For the purpose of simulating traffic flow in pluvial flooding scenarios,effects of rain were fully concerned during the design of routing algorithm,car-following model and road cross model,which were three key components in the traffic simulation method.(3)Develop an easy tool to implement dynamic traffic simulation in pluvial flooding scenarios.The traffic simulation tool was developed from the bottom up using C#based on GIS and Microsoft Visual Studio 2010.Its functions and data requirements were well designed according to the traffic simulation method established in the last step.Specifically,if one input a topology of road network,trip origins and destinations(OD)and precipitation data,it can implement trip assignment(to find routes in the network)and dynamic simulation traffic of flow,and then output the travel time,distance and the vehicles counted on each road segment.(4)Evaluate the effects of pluvial flooding on urban traffic through a case study facilitated by the established simulation method and tool.As another important step,traffic modeling results were validated before doing analysis for the case study.Specifically,through a comparison with the travel routes planed by GoogleMaps,and traffic volume data collected from two real road crosses,the reliability of simulated routes and traffic volume were proved to be adequate for further evaluation.Based on simulation results of 12 compound scenarios,it was clear that pluvial flooding may affect urban traffic from several aspects.First,compared with the sunny weather,average travel time and distance in the rainy weather would increase due to road closures and a decline of vehicle speed.Second,with the magnitude of precipitation increasing,more extensive road closures would last for longer hours,which consequently exacerbate the travel delay.Third,the change of traffic volume on most road segments was not obvious,but it exhibited an evident response to pluvial flooding about 30 min later than the spatiotemporal change of road condition over the entire network.